The Indonesian archipelago: an ancient genetic highway linking Asia and the Pacific

ORIGINAL ARTICLE

The Indonesian archipelago: an ancient genetichighway linking Asia and the Pacific

Meryanne K Tumonggor1,2, Tatiana M Karafet3, Brian Hallmark3, J Stephen Lansing1,4, Herawati Sudoyo2,Michael F Hammer3 and Murray P Cox5

Indonesia, an island nation linking mainland Asia with the Pacific world, hosts a wide range of linguistic, ethnic and genetic

diversity. Despite the complexity of this cultural environment, genetic studies in Indonesia remain surprisingly sparse. Here,

we report mitochondrial DNA (mtDNA) and associated Y-chromosome diversity for the largest cohort of Indonesians examined

to date—2740 individuals from 70 communities spanning 12 islands across the breadth of the Indonesian archipelago.

We reconstruct 50 000 years of population movements, from mitochondrial lineages reflecting the very earliest settlers in island

southeast Asia, to Neolithic population dispersals. Historic contacts from Chinese, Indians, Arabs and Europeans comprise

a noticeable fraction of Y-chromosome variation, but are not reflected in the maternally inherited mtDNA. While this historic

immigration favored men, patterns of genetic diversity show that women moved more widely in earlier times. However, measures

of population differentiation signal that Indonesian communities are trending away from the matri- or ambilocality of early

Austronesian societies toward the more common practice of patrilocal residence today. Such sex-specific dispersal patterns

remain even after correcting for the different mutation rates of mtDNA and the Y chromosome. This detailed palimpsest of

Indonesian genetic diversity is a direct outcome of the region’s complex history of immigration, transitory migrants and

populations that have endured in situ since the region’s first settlement.

Journal of Human Genetics (2013) 58, 165–173; doi:10.1038/jhg.2012.154; published online 24 January 2013

Keywords: Indonesia; mitochondrial DNA; molecular anthropology; Y chromosome

INTRODUCTION

Indonesia, a maritime nation comprising over 17 000 islands strad-dling the Pacific and Indian Oceans, links mainland Asia with thePacific world. Although a single lingua franca is spoken widely acrossthe archipelago today (Bahasa Indonesia), Indonesia hosts over 730indigenous languages and associated ethnic groups.1 Most of theselanguages belong to the geographically dispersed Austronesianlanguage family, but Papuan languages are spoken by some groupsin the far east of the archipelago.2 Relative to its land area, Indonesiais one of the most varied regions on earth in terms of ethnic,linguistic and genetic diversity.

As an island nation, past changes in global climate have had anespecially strong influence on Indonesia.3 Lands in the west, nowlargely submerged, once formed a vast continental shelf jutting outfrom Asia (Sundaland). In the east, Australia and New Guinea werelinked into a single continent (Sahul). During most of the latePleistocene, it was possible to walk—with only minor watercrossings—from Bangkok to Sydney. However, following the end ofthe last glacial period, B18 kya, ice melting in the arctic fueled a rapidrise in global sea levels. Continental Indonesia swiftly fragmented intothe long chain of islands that characterizes the nation today.

The human history of Indonesia played out against the backdrop ofthis dynamically changing geography. Settled by anatomically modernhumans at least 47 kya4,5 and perhaps much earlier,6 our species hasinhabited Indonesia longer than Europe.7,8 Archeology providessporadic, but widespread, evidence of early hunter-gatherer groupsthroughout the Pleistocene.9 However, the biggest cultural changeoccurred within the last 10 kya when the archeological record abruptlychronicles the appearance of agricultural communities together withpottery, plant cultivation and animal domestication. The vast spreadof Austronesian languages likely also occurred during this time.10–12

Whether the Neolithic era was ushered in by population move-ments from Taiwan13 or was instead dominated by regional develop-ments14,15 remains a highly contentious topic of discussion—a question that genetics is increasingly striving to answer.

For such a large and ethnically diverse nation at the pivot point ofAsia and the Pacific, studies of Indonesian genetic diversity aresurprisingly sparse.16–24 The most complete study of mitochondrialDNA (mtDNA) posited that Indonesian diversity has largely beenshaped by two forces: population movements driven by sea levelchanges, and by farming populations expanding from the Asianmainland into the islands of southeast Asia.25,26 Y-chromosome

1Department of Anthropology, University of Arizona, Tucson, AZ, USA; 2Eijkman Institute for Molecular Biology, Jakarta, Indonesia; 3Arizona Research Laboratories, Division ofBiotechnology, University of Arizona, Tucson, AZ, USA; 4Santa Fe Institute, Santa Fe, NM, USA and 5Institute of Fundamental Sciences, Massey University, Palmerston North,New ZealandCorrespondence: Professor MF Hammer, Arizona Research Laboratories, Division of Biotechnology, University of Arizona, Tucson, AZ 85721, USA.E-mail: [email protected]

Received 23 September 2012; revised 28 November 2012; accepted 14 December 2012; published online 24 January 2013

Journal of Human Genetics (2013) 58, 165–173& 2013 The Japan Society of Human Genetics All rights reserved 1434-5161/13

www.nature.com/jhg

http://dx.doi.org/10.1038/jhg.2012.154

mailto:[email protected]

http://www.nature.com/jhg

studies, reflecting the history of Indonesian men, mirror these themes,emphasizing a complex multifaceted history of the region’s islandsand communities.20 Although autosomal markers are rapidlychanging our understanding of regional prehistory,27–29 thegeographical resolution and widespread availability of comparativedata ensure that haploid markers remain extremely powerful tools forstudying the human past. Here, we present the largest sample ofIndonesian mtDNA diversity assembled to date—2740 individualsfrom 70 populations on 12 islands spanning the full range ofIndonesia’s geographical, ethnic and linguistic diversity. We recon-struct a partial history of Indonesia’s women, piecing together 50 ky ofpopulation movements that have shaped the diversity of Indonesiansliving today. Finally, by comparison with Y-chromosome data for thesame set of individuals,20 we contrast the histories of Indonesian menand women, presenting common patterns of shared inheritance withkey points of demographic and social difference.

MATERIALS AND METHODS

SamplesGenetic diversity was screened in 2740 consenting, closely unrelated and

seemingly healthy individuals drawn from across the Indonesian archipelago.

Permission to conduct research in Indonesia was granted by the Indonesian

Institute of Sciences. Biological samples (peripheral blood and buccal swabs)

were obtained by MKT, JSL, HS, Golfiani Malik, Wuryantari Setiadi and Loa

Helena Suryadi of the Eijkman Institute for Molecular Biology, Jakarta,

Indonesia, with the assistance of Indonesian Public Health clinic staff, and

followed protocols for the protection of human subjects established by both

the Eijkman Institute and the University of Arizona Institutional Review

Boards. Participant interviews confirmed ethnic, linguistic and geographic

classifications for at least two generations into the past.

Seventy populations from 12 island groups were sampled across the

Indonesian archipelago, including (from west to east) Sumatra (n¼ 42),

Mentawai (n¼ 128), Nias (n¼ 59), Java (n¼ 51), Bali (n¼ 487), Sulawesi

(n¼ 200), Sumba (n¼ 634), Flores (n¼ 469), Lembata (n¼ 92), Pantar

(n¼ 29), Timor (n¼ 526) and Alor (n¼ 23) (Supplementary Table S1). For

comparative purposes, we also explored previously published genetic data from

mainland Asian populations: northwestern Chinese (Yili, Xinjiang; n¼ 47),

northeastern Chinese (Fencheng, Liaoning; n¼ 51), southwestern Chinese

(Kunming, Yunnan; n¼ 43) and southeastern Chinese (Zhanjiang, Guang-

dong; n¼ 30),30 Thai (n¼ 52) and Vietnamese (n¼ 41),31,32 and indigenous

Malaysians (n¼ 260).33 We also compared our Indonesian data with

neighboring island southeast Asian populations, including indigenous

Taiwanese (n¼ 640),34 Filipinos (n¼ 423), Papua New Guineans (n¼ 231),

Island Melanesians (n¼ 1366) and Micronesians (n¼ 47).35,36 The geo-

graphical locations of these populations are illustrated in Figure 1.

DNA extraction and genetic screeningDNA was extracted from peripheral blood samples using the salting-out

procedure of Miller, Dykes and Polesky.37 DNA from buccal swabs was

extracted using standard phenol–chloroform protocols.

The first hypervariable segment (HVS I) of mtDNA was amplified using

primers L15926 (50-TCAAAGCTTACACCAGTCTTGTAAACC-30) and H639

(50-GGGTGATGTGAGCCCGTCA-30). PCR amplicons were sequenced in both

forward and reverse directions using primers L15965 (50-CAAGGACAAAT

CAGAGAA-30) and H11 (50-GTGGTTAATAGGGTGATAG-30). Traditional

Sanger sequences were aligned and edited with Sequencher v. 5.0 (Gene Codes

Corporation, Ann Arbor, MI, USA; http://www.genecodes.com). Polymor-

phisms were scored relative to the revised Cambridge reference sequence.38

Haplogroups were initially predicted from HVS I sequences using known

reference genomes.23,25,36,39–42 Assignments were subsequently confirmed using

Taqman and restriction fragment length polymorphism assays (Supplementary

Table S2).

Figure 1 Locations of Studied Populations. (1) Northwest China,30 (2) northeast China,30 (3) southwest China,30 (4) southeast China,30 (5) Indigenous

Taiwanese,34 (6) Thailand,31 (7) Vietnam,32 (8) Philippines,36 (9) Indigenous Malaysians,33 (10) Sumatra (present study), (11) Nias (present study), (12)

Mentawai (present study), (13) Java (present study), (14) Bali (present study), (15) Sumba (present study), (16) Flores (present study), (17) Lembata

(present study), (18) Pantar (present study), (19) Timor (present study), (20) Alor (present study), (21) Sulawesi (present study), (22) Papua New Guinea,35

(23) Micronesia35 and (24) Melanesia35.

Indonesian genetic highwayMK Tumonggor et al

166

Journal of Human Genetics

http://www.genecodes.com

Statistical analysesMolecular diversity, population structure estimates and genetic distances

between populations were calculated using Arlequin v. 3.1143 (http://cmpg.

unibe.ch/software/arlequin3). The statistical significance of FST values was

evaluated using 104 permutations of the data. Pairwise genetic distances

between populations were computed as the linearized value, FST/(1�FST).44,45

Differences in haplogroup frequencies between eastern and western Indonesia

were determined via parametric bootstrapping using code implemented in

R (available from the authors on request).46

Median-joining networks were built using Network v. 4.5.1.6 (Fluxus

Engineering; http://www.fluxus-engineering.com).47 Haplogroups were tenta-

tively dated with the r statistic method48 using a rate of one mutation every

19 171 years.49 Dates are only intended as a rough guide for relative

haplogroup ages.50

MtDNA and Y-chromosome comparisonsY-chromosome diversity has also been screened for the same panel of

individuals described above.20 Differences in mtDNA and Y-chromosome

diversity between populations were analyzed using an analysis of molecular

variance implemented in Arlequin. A measure of interlocus differentiation

GST,51 standardized for different mutation rates, was calculated using code

implemented in R (available from the authors on request).46

MtDNA HVS I sequences have been deposited in GenBank (accession

numbers: KC113641–KC115854). Y-chromosome STR data are provided as

Supplementary Data Set S1.

RESULTS

We screened 2740 individuals from 70 communities on 12 Indonesianislands. MtDNA HVS I sequences showed high levels of diversity, asmeasured by the number of polymorphic sites, number of haplotypes,the mean number of pairwise differences and Nei’s haplotype diversity(Table 1). The haplotype diversity of communities ranged from0.862–0.996, which indicates that most individuals within thesegroups carry unique mtDNA lineages. The most diverse communitiesare found in eastern parts of the archipelago (Sumba, Flores, Pantarand Alor), where both Asian and Papuan lineages occur side by side.The western barrier islands of Nias and Mentawai are least diverse,even when compared with other Asian populations. For instance, Niasand Mentawai are the only Indonesian populations with diversity aslow as that of indigenous Taiwanese groups (0.838–0.924).34,52

Summary statistics such as Fu’s Fs and Tajima’s D can beinformative about the roles of selection and demography. For mtDNAcontrol region sequences, which seem little affected by naturalselection, observed values are suggestive of low levels of growthacross the archipelago, with the exception of the barrier islands, Niasand Mentawai (Table 1). Growth seems strongest in the central clusterof islands (Bali, Sulawesi, Sumba and Flores), but is less pronouncedin the extreme west and east of the archipelago.

Individuals were assigned to mtDNA haplogroups using a combi-nation of HVS I sequence motifs and single-nucleotide polymor-phisms (SNPs) distributed around the coding region of the mtDNAgenome. Fifty-one haplogroups were identified, with all lineagesfalling into macrohaplogroups M (47.05%) and N (52.95%). The51 Indonesian haplogroups are plotted on a tree of mtDNA diversityconstructed using previously published HVS I sequences and coding-region SNPs23,25,36,39–42 (Figure 2).

Haplogroup frequencies differ between western and easternIndonesia (Supplementary Table S3). In the west, haplogroups B5a(12%), B4c1b3 (9%) and Y2 (10.5%) are carried by a third ofindividuals. These haplogroups, frequent in western Indonesia, arenotable by their near absence in eastern Indonesia. In the east,haplogroups F1a4 (8.7%), Q including Q1 and Q2 (7.7%), P (2.8%)and B4a1a1a (2.3%) represent nearly a quarter of individuals.Correspondingly, these haplogroups are rare or absent in westernIndonesia, which is expected for lineages with strong Papuanconnections (P and Q), but more surprising for lineages like thePolynesian motif (B4a1a1a). The Polynesian motif is found as far westas Bali, albeit in just two individuals (0.4%). However, it was notdetected in samples from the western Indonesian islands of Java,Sumatra, Nias and Mentawai, even though this region is thought tohave contributed to the settlement of Madagascar where thePolynesian motif is carried by nearly a third of individuals.53,54 Theprevalence of the Polynesian motif in Madagascar, and its absencefrom the island region where the inhabitants of Madagascaroriginated, has yet to be satisfactorily explained.

We compared the distribution of Indonesian mtDNA haplogroupswith those of surrounding populations (Supplementary Table S4).Most haplogroups are shared. The deep maternal lineages M17a, M73,M47, N21, N22, R21, R22 and R23 have patchy distributions across

Table 1 Molecular diversity indices and growth summary statistics for Indonesian island groups

Diversity indices Growth Statistics

Island N h S Haplotype diversity s.d. MNPD s.d. Nucleotide diversity s.d. Tajima’s D P Fu’s Fs P

Sumatra 42 29 48 0.973 0.013 7.90 3.75 0.015 0.008 �1.03 0.146 �13.2 o0.001

Mentawai 128 20 38 0.890 0.012 6.77 3.21 0.013 0.007 �0.10 0.504 0.500 0.632

Nias 59 24 44 0.862 0.040 7.02 3.34 0.013 0.007 �0.87 0.196 �4.54 0.104

Java 51 33 52 0.968 0.013 7.68 3.64 0.014 0.008 �1.16 0.084 �16.1 o0.001

Bali 487 129 99 0.975 0.002 8.16 3.79 0.015 0.008 �1.29 0.071 �24.1 0.002

Sulawesi 200 97 87 0.976 0.005 7.53 3.53 0.014 0.007 �1.51 0.040 �24.5 0.001

Sumba 634 159 108 0.981 0.001 7.89 3.68 0.015 0.008 �1.40 0.038 �24.0 0.001

Flores 469 149 109 0.987 0.001 8.22 3.82 0.016 0.008 �1.44 0.038 �24.1 0.002

Lembata 92 47 68 0.968 0.008 9.12 4.23 0.017 0.009 �1.04 0.154 �20.7 o0.001

Timor 526 117 96 0.955 0.005 8.72 4.03 0.016 0.008 �1.08 0.115 �24.0 0.002

Alor 23 22 45 0.996 0.014 9.12 4.36 0.017 0.009 �0.98 0.176 �13.4 o0.001

Pantar 29 22 52 0.978 0.015 10.0 4.73 0.019 0.010 �0.89 0.192 �6.78 0.016

Abbreviations: h, number of haplotypes; MNPD, mean number of pairwise differences; N, number of sequences; P, probability value; S, number of polymorphic sites.Significant growth summary statistics are bold and italicized.


167


http://cmpg.unibe.ch/software/arlequin3

http://cmpg.unibe.ch/software/arlequin3

http://www.fluxus-engineering.com

mainland and island southeast Asia, likely reflecting ancient maternallineages tracing back to the first settlers in this region.25,33,36,55 Four ofthese lineages (M17a, N21, R22 and R23) reach higher frequencies inwestern compared with eastern Indonesia (parametric bootstrap, allP{0.01) (Supplementary Table S3).

Owing to proposed population origins of Austronesian languagespeakers in Taiwan, Indonesian links to the Philippines and indigen-ous Taiwanese are of especial interest. These three locations share fourhaplogroups (E1a1a, M7b3, M7c3c and Y2), which have previouslybeen suggested as candidates for a mid-Holocene dispersal out ofTaiwan.25,36 Figure 3 illustrates that lineages shared with Filipinos andindigenous Taiwanese are generally more common in the east than inthe west of Indonesia. The exceptions are Y2 and M7c3c, where highfrequencies in Nias and Mentawai may perhaps be caused by geneticdrift or strong founder events in these extremely small and geogra-phically isolated populations on the barrier islands of Sumatra.52

Median-joining networks were constructed for haplogroups sharedbetween Indonesia, the Philippines and Taiwan (Figure 3). We notethat at least one haplotype was shared between Indonesia and Taiwanfor each of these four lineages (E1a1a, M7b3, M7c3c and Y2). In allcases, this shared lineage was the ancestral haplotype, and descendentlineages depict a star-like expansion indicative of population growthand/or geographical expansion. Unfortunately, the networks are notinformative about the direction of migration: the data would fit amodel of rapid expansion from Taiwan to the Philippines andIndonesia, but are equally consistent with population movements inthe opposite direction.

To explore population relationships further, multidimensionalscaling was performed on all Indonesian HVS I sequences usingSlatkin’s linearized FST as the genetic distance between groups(Supplementary Figure S1). Although there is no simple population

division, western Indonesian groups cluster away from easternIndonesian populations. Multidimensional scaling analysis withregional neighbors shows that Indonesians fall together with Asiangroups, but away from Oceanian populations (Supplementary FigureS2). Among Asian populations, Indonesians cluster most closely withthe Philippines and Vietnam, and more distantly with Taiwan(Supplementary Figure S3).

An analysis of molecular variance illustrates that the Y-chromo-some STRs (FST¼ 0.202) have markedly higher variation amongpopulations than mtDNA HVS I (FST¼ 0.073) (Table 2), suggestingthat women have dispersed more widely in the past than men. Thistrend is maintained when populations are collapsed to their 12 islandgroups or an even broader east–west division. The trend holds, albeitmore weakly, when distances are standardized for the B400-foldhigher mutation rate of the Y chromosome (on the order of 10�5

mutation events per STR per year)56–59 relative to mtDNA (on theorder of 10�7 mutation events per base pair per year).49 The GST ofY-chromosome STRs (GST¼ 0.972) is still notably higher than that ofmtDNA HVS I (GST¼ 0.862) (Supplementary Table S5). Curiously,the trend even holds on very small geographical scales, such as amongcommunities situated along the highland river systems of Bali.60

DISCUSSION

Austronesian languages, spoken from Madagascar in the west toRapanui/Easter Island in the east, form one of the world’s largest andmost geographically dispersed language families. Despite intensearcheological, linguistic and genetic research, where people speakingthese languages came from, how they dispersed and what theirancestral communities looked like remain major open questions. Onehypothesis states that Austronesian speakers originated in Taiwan,which is supported by the fact that the most basal languages of the

Figure 2 Phylogeny of mitochondrial DNA haplogroups observed in Indonesian populations in the present study.


168


Figure 3 Frequency distributions and median-joining networks of mitochondrial DNA haplogroups (a) M7b3, (b) E1a1a, (c) Y2 and (d) M7c3c shared by

indigenous Taiwanese (green), Filipino (blue) and Indonesian populations (yellow). Note that haplogroup networks are largely agnostic about the direction of

population movements between Taiwan, the Philippines and Indonesia.

Table 2 Analysis of molecular variance (AMOVA) for subsets of Indonesian populations

Within populations Among populations within groups Among groups

FST Var (%) FSC Var (%) FCT Var (%)

Y-SNPs

All Indonesian populations 70 populations 0.361 63.9

Islands 12 islands 0.384 61.6 0.195 14.9 0.236 23.5

East vs West 2 main groups 0.475 52.5 0.247 17.2 0.304 30.3

Y-STRs


Islands 12 islands 0.210 78.9 0.152 14.2 0.069 6.90


mtDNA SNPs


Islands 12 islands 0.096 90.4 0.067 6.45 0.032 3.15


mtDNA HVS I


Islands 12 islands 0.077 92.3 0.048 4.61 0.031 3.09


Abbreviations: HVS I, first hypervariable segment; mtDNA, mitochondrial DNA; SNP, single-nucleotide polymorphism; Var, variance.All values are statistically significant (P{0.0001).


169


Austronesian family are spoken there.61 From Taiwan, nascentfarming groups are believed to have expanded 3–4000 years agothrough the Philippines, Indonesia and out into the Pacific.3,13

Alternative hypotheses, such as Austronesian groups originating inand dispersing from Indonesia, remain possible,14,62 with manygenetic lineages in Indonesia showing old and local connections.Models combining these two extremes may ultimately be the bestpredictors. We envisage some genetic contributions from Taiwan,possibly including speakers of early Austronesian languages, with asubstantial biological heritage from waves of ancestral populationsarriving in island southeast Asia following its first settlement 50 kya.

We consider how the mitochondrial data aligns with this spectrumof origin models. Some older lineages appear to trace back to the veryearliest settlers in southeast Asia. R21, which is found only inMentawai (0.8%), diverged from the common haplogroup R ancestorB60 kya.55,63 Other basal mtDNA lineages (for example, N21 andN22) are shared at extremely low levels by Indonesians and mainlandsoutheast Asian groups.

Candidate mtDNA lineages of a Taiwanese dispersal (E1a1a, M7b3,M7c3c and Y2) have also been proposed.25,36 These four haplogroupshave similar distributions, with basal haplotypes shared betweenindigenous Taiwanese, Filipinos and Indonesians. However, thedirection of dispersal is inconclusive: ancestral-derived haplo-type orders are consistent with a rapid expansion from Taiwan tothe Philippines and Indonesia, but population dispersals in theopposite direction are equally likely. Although we providehaplogroup dates with some reluctance,50 we note that relative agesare inconsistent with a simple dispersal from Taiwan to thePhilippines, and thence to Indonesia. Instead, they seem a better fitto widespread population movements within island southeast Asiaduring the Holocene. However, we note that sample sizes differsubstantially between these three locations (2740, 423 and 640 forIndonesians, Filipinos and indigenous Taiwanese, respectively), whichadds variance—and perhaps bias—to diversity and dating estimates.Considered together with their large confidence intervals, we arereluctant to draw strong conclusions from molecular dates alone.

The Polynesian motif is also generally associated with a Taiwanesedispersal, but actually possesses an unusual geographical distribution.The ancestral form occurs widely throughout mainland and islandsoutheast Asia. However, the Polynesian motif itself is found only atlow frequency in the Philippines (0.5%)36 and eastern Indonesia(2.3%).16 Although frequencies reach as high as 7.4% on Timor, thelineage is found no further west than Bali (0.4%, or just 2 of 457individuals). This is consistent with a proposed origin in islandMelanesia,64 but notably conflicts with the high frequency of thePolynesian motif in Madagascar, which was settled B1200 years agofrom western Indonesian sources.53 We suggest that an inclusiveframework that describes the full distribution of this unusual mtDNAlineage is still lacking. Nevertheless, an unambiguous connection withpopulation dispersals from Taiwan during the Neolithic seemsincreasingly unlikely.

MtDNA evidence does suggest that most Indonesian groups mayhave increased in size. Negative values of Fu’s Fs and Tajima’s D pointtoward population growth and/or geographical expansion (Table 1),and reinforce similar conclusions that might be drawn from the star-like phylogenies of shared island southeast Asian mtDNA lineages(Figure 3). Signals of growth are greatest in the center of thearchipelago (Bali, Sulawesi, Sumba and Flores), but weaker towardsthe eastern and western peripheries. Key exceptions are the barrierislands, Nias and Mentawai, whose mtDNA profiles are statisticallymore consistent with constant population size. These two groups also

show some of the lowest haplotype diversities of any island southeastAsian populations (for instance, 43.5% of the Nias population carryhaplogroup Y2). These outlier patterns may be due to genetic drift, assettlements on Nias and Mentawai are small even by indigenousIndonesian standards, or they may be caused by founder events andtherefore reflect the unusual genetic profile of the islands’ first settlers.The strong retained cultural heritage of these barrier islands mayspeak to this point. The inhabitants of Nias still practice oldtraditions, including the construction and maintenance of megaliths,and stone jumping (hombo batu), whereby young men showprowess by repeatedly jumping up onto a tall stone. Megalithictraditions associated with Austronesian culture were once commonacross Indonesia, but are now largely restricted to peripheral com-munities including those on Nias, Mentawai and Sumba.

In comparisons with neighboring populations, Indonesia’s closestgenetic connections lie toward mainland and island southeast Asiarather than Oceania (Supplementary Figures S2 and S3). WesternIndonesian groups are notably distinct from Papuan groups(Figure 4), largely owing to low levels of haplogroups P and Q.65

Whether these lineages are an enduring local presence from theoriginal inhabitants of the region, or instead reflect recent westwardmovements from New Guinea is unclear. Soares et al.64 suggest thatthe Polynesian motif spread westward from the Bismarck Archipelago,and similar movements have been proposed to explain thedistribution of Papuan languages in eastern Indonesia.2,66 SNPsfrom across the genome have been used to argue for an indigenouspresence of Papuan genotypes rather than back migration from NewGuinea, but strong statistical support for this is currently lacking.67

Unfortunately, the data presented here, while powerful in helping todistinguish shared connections, are less informative about directionsof movement.

The individuals screened for mtDNA were also assayed forY-chromosome diversity.20 Analysis of Austronesian languages andcultural systems,10,11,68 as well as autosomal markers,27,29 suggests thatthe men and women of island southeast Asia have followed quitedifferent social histories. We compared mtDNA and Y-chromosomediversity to explore this further. Genetic divisions betweenpopulations are far weaker for mtDNA HVS I (FST¼ 0.073) thanfor Y-chromosome STRs (FST¼ 0.202), and this effect is even morepronounced at the haplogroup level when Indonesia is separatedinto its eastern and western parts (mtDNA SNPs FST¼ 0.109;Y-chromosome SNPs FST¼ 0.475) (Table 2). This discrepancy sug-gests that men and women have had different patterns of dispersal,with women moving widely between communities, while men havehistorically stayed local. One possible social explanation is patrilo-cality, where men remain in their natal community, but women moveto the home village of their husband. Indeed, most of the populationspresented in this study are patrilineal today. Interestingly, matrilinealsystems are thought to have dominated ancestral Austronesiansocieties.11,27,69 These mtDNA/Y-chromosome patterns may havebeen laid down after that cultural shift, and therefore reflect onlythe last few thousand years. Alternately, patrilocality could insteadbe the long-term standard with a transient switch to matrilocalityduring the Austronesian era. Differences in mutation rates couldpotentially confound this analysis; Y-STRs have high mutation rates,even when compared with the fast evolving mtDNA. However, thesegeneral patterns are still supported even when using standardizedmeasures of genetic variation (Supplementary Table S5).

However, while the genetic evidence indicates that men and womenexperienced divergent histories, shared characteristics are morepronounced than differences. Karafet et al.20 proposed a four-stage


170


colonization model built around variation on the Y chromosome.Here, we integrate the mtDNA evidence to provide a broaderframework for reconstructing the history of Indonesia.

The first stage of Indonesian prehistory represents the archipelago’sinitial settlement as part of the African dispersal B50 kya. Thegeography of the region was then markedly different from today. Sealevels were much lower, most modern islands had merged into largerlandmasses and the westernmost parts of Indonesia were physicallycontiguous with mainland Asia. This first stage is recorded by deepmtDNA lineages (M17a, M73, M47, N21, N22, R21, R22 and R23),which trace back to the main branching of macrohaplogroups M andN, and have a spotty distribution across both mainland and islandsoutheast Asia today.25,33,36,55

The second stage reflects recurrent colonization events frommainland Asia throughout the later Paleolithic. Many haplogroups

(B4a, B4b, B4c, B4c1b3, B5a, B5b, B5b1, D and E) show origin datesof 10–40 kya (Supplementary Table S6)25,63 and are distributed acrossa wide range of mainland and island southeast Asian populations.As these lineages vary considerably in diversity (and hence, probableage) and show quite different geographical distributions, it is unlikelythat any single demic event brought them to Indonesia. Indeed, manyof these mtDNA haplogroups have been identified as key componentsof populations in peninsular Malaysia.63 Therefore these lineageslikely reflect multiple population movements from mainland Asia,possibly hunter-gatherers who followed the now-submerged riversystems that once ran from mainland Asia between the modernislands of Sumatra, Java and Borneo.

The third stage represents Neolithic movements into and aroundisland southeast Asia. Some of these may involve populationdispersals from (and perhaps to) Taiwan, while others reflect

Figure 4 Frequency distribution of mitochondrial DNA haplogroups (a) B4a1a1a, (b) Q, (c) P and (d) M7b1.


171


movements between Indonesian island groups. Representative hap-logroups include M7b3, E1a1a, M7c3c and Y2. Autosomal datastrongly support large demic movements of Asian populations intoeastern Indonesia from around 4 kya.67 Putatively associated with thespread of Austronesian languages, the direction of these movementsremains unknown. If population movements during the Holocenewere largely restricted to eastern Indonesia, this may explain whybasal mtDNA lineages (such as M17a, N21, R22 and R23) are muchrarer in the east. Although the Neolithic seems to have been a periodof high population movement, it is curious that distinctions betweeneastern and western parts of the archipelago persist to the present.Within these regions, genetic similarity between island groupssuggests that population movements were relatively frequent,consistent with archeological and linguistic evidence of a strongmaritime culture. Borders such as Wallace’s biogeographical line arepoor boundaries for communities with significant ocean-goingcapabilities,17 and the reason for an east–west distinction so clearlypersisting in the genetic data is still imperfectly understood.17,27,70 Themost inclusive explanations invoke changes in social behavior, suchas the emergence of matrilocal ‘house societies’ (societes a maison)during the early Austronesian period, and their subsequentdisappearance as communities increasingly turned to patrilocalitywhen the expansion period drew to a close.70

The fourth stage reflects historic movements into Indonesia, largelyinvolving trade and the associated spread of major religions fromIndia, Arabia and China.71 Although found at relatively low frequencytoday, Y-chromosome lineages representing these movements occuracross Indonesia,20 notably in the west, such as the Hindu dominatedisland of Bali.21 It is therefore interesting that no mtDNA lineagesrepresenting this period of invasion and migration were observed inour substantial data set. Unlike previous stages of Indonesiansettlement, trade and religious connections during the historic eramust have involved only male travelers, who subsequently took localwives. This difference between mtDNA and the Y chromosomeemphasizes the complex, plural nature of most social processes.While most indigenous Indonesian communities practice patrilocalitytoday (and hence exhibit preferential movement of women), long-distance genetic contributions are still effectively driven by men.As is often the case with complex biological systems, seeminglydiametrically opposed processes can even act simultaneously in thesame community at the same time.

The enormous diversity of language, culture and genetics inIndonesia is a direct outcome of the region’s complex history ofmigration and settlement. Demic and cultural processes are bothapparent: movements of people are indicated by related mtDNAlineages, but many aspects of culture—notably the widespreaddispersal of Austronesian languages—are not obviously associatedwith genetics. Indeed, the only mtDNA lineage found across allIndonesian island groups is M7c3c, but this haplogroup, while alsopresent in Taiwan and the Philippines, appears to be absent fromother Austronesian-speaking populations in Oceania. Therefore, nosingle shared mtDNA lineage links all speakers of Austronesianlanguages, even if only at low frequency. Instead, Austronesianpopulations are characterized more by their diversity than by anyshared genetic inheritance. However, there are limits to the power ofuniparentally inherited markers, and with nearly 3000 individualsnow screened, we wonder whether further sampling will substantiallychange the picture of mtDNA diversity portrayed here. Admixtureanalysis on genome-wide data sets indicates the extent of Asianimmigrants in Indonesian populations and assigns the time ofadmixture to the mid to late Holocene.67 Therefore, we are hope-

ful that new studies of autosomal data may help to answer manyof the questions that remain outstanding, and we look forward tothe clarification that this new wave of genetic evidence promisesto bring.

ACKNOWLEDGEMENTSThis research was supported by a US National Science Foundation grant (SES

0725470) to JSL, MFH, TMK, and Joe C. Watkins, which funded the doctoral

research of MKT. The Royal Society of New Zealand provided support

through a Rutherford Fellowship (RDF-10-MAU-001) and Marsden Grant

(11-MAU-007) to MPC.

1 Lewis, M. P. Ethnologue: Languages of the World (SIL International, Dallas, TX, USA,2009).

2 Pawley, A. The chequered career of the Trans New Guinea hypothesis: Recent researchand its implications. In Papuan Pasts: Cultural, Linguistic and Biological Histories ofPapuan-Speaking Peoples (eds Pawley, A., Attenborough, R., Golson, J. & Hide, R.)67–107 (Pacific Linguistics, Canberra, 2005).

3 Bellwood, P. Prehistory of the Indo-Malaysian Archipelago (Australian NationalUniversity, Canberra, 2007).

4 Barker, G. The archaeology of foraging and farming at Niah Cave, Sarawak. AsianPerspect. 44, 90–106 (2005).

5 Barker, G., Barton, H., Bird, M., Daly, P., Datan, I., Dykes, A. et al. The ‘humanrevolution’ in lowland tropical Southeast Asia: the antiquity and behavior of anato-mically modern humans at Niah Cave (Sarawak, Borneo). J. Hum. Evol. 52, 243–261(2007).

6 Rasmussen, M., Guo, X., Wang, Y., Lohmueller, K. E., Rasmussen, S., Albrechtsen, A.et al. An Aboriginal Australian genome reveals separate human dispersals into Asia.Science 334, 94–98 (2011).

7 Mellars, P. A new radiocarbon revolution and the dispersal of modern humans inEurasia. Nature 439, 931–935 (2006).

8 Mellars, P. Going east: new genetic and archaeological perspectives on the modernhuman colonization of Eurasia. Science 313, 796–800 (2006).

9 O’Connell, J. F. & Allen, J. Dating the colonization of Sahul (PleistoceneAustralia–New Guinea): a review of recent research. J. Archaeol. Sci. 31, 835–853(2004).

10 Gray, R. D., Drummond, A. J. & Greenhill, S. J. Language phylogenies reveal expansionpulses and pauses in Pacific settlement. Science 323, 479–483 (2009).

11 Jordan, F. M., Gray, R. D., Greenhill, S. J. & Mace, R. Matrilocal residence is ancestralin Austronesian societies. Proc. R. Soc. B 276, 1957–1964 (2009).

12 Currie, T. E., Greenhill, S. J., Gray, R. D., Hasegawa, T. & Mace, R. Rise and fall ofpolitical complexity in island south-east Asia and the Pacific. Nature 467, 801–804(2010).

13 Bellwood, P. The First Farmers: The Origins of Agricultural Societies (BlackwellPublishing, Oxford, 2005).

14 Oppenheimer, S. Eden in the East: The Drowned Continent of Southeast Asia (Phoenix,London, 1999).

15 Oppenheimer, S. J. & Richards, M. Polynesian origins. Slow boat to Melanesia? Nature410, 166–167 (2001).

16 Cox, M. P. Indonesian mitochondrial DNA and its opposition to a Pleistocene era originof proto-Polynesians in Island Southeast Asia. Hum. Biol. 77, 179–188 (2005).

17 Cox, M. P. The genetic environment of Melanesia: Clines, clusters and contact. InPopulation Genetics Research Progress (ed. Koven, V. T.) 45–83 (Nova SciencePublishers, New York, 2008).

18 Cox, M. P. & Lahr, M. M. Y-Chromosome diversity is inversely associated with languageaffiliation in paired Austronesian- and Papuan-speaking communities from SolomonIslands. Am. J. Hum. Biol. 18, 35–50 (2006).

19 Cox, M. P., Redd, A. J., Karafet, T. M., Ponder, C. A., Lansing, J. S., Sudoyo, H. et al. APolynesian motif on the Y chromosome: population structure in Remote Oceania. Hum.Biol. 79, 525–535 (2007).

20 Karafet, T. M., Hallmark, B., Cox, M. P., Sudoyo, H., Downey, S. S., Lansing, J. S. et al.Major east-west division underlies Y chromosome stratification across Indonesia. Mol.Biol. Evol. 27, 1833–1844 (2010).

21 Karafet, T. M., Lansing, J. S., Redd, A. J., Reznikova, S., Watkins, J. C., Surata, S. P.et al. Balinese Y-chromosome perspective on the peopling of Indonesia: geneticcontributions from pre-Neolithic hunter-gatherers, Austronesian farmers, and Indiantraders. Hum. Biol. 77, 93–114 (2005).

22 Kayser, M. The human genetic history of Oceania: near and remote views of dispersal.Curr. Biol. 20, R194–R201 (2010).

23 Kayser, M., Brauer, S., Cordaux, R., Casto, A., Lao, O., Zhivotovsky, L. A. et al.Melanesian and Asian origins of Polynesians: mtDNA and Y-chromosome gradientsacross the Pacific. Mol. Biol. Evol. 23, 2234–2244 (2006).

24 Kayser, M., Brauer, S., Weiss, G., Schiefenhovel, W., Underhill, P., Shen, P. et al.Reduced Y-chromosome, but not mitochondrial DNA, diversity in human populationsfrom West New Guinea. Am. J. Hum. Genet. 72, 281–302 (2003).


172


25 Hill, C., Soares, P., Mormina, M., Macaulay, V., Clarke, D., Blumbach, P. B. et al. Amitochondrial stratigraphy for island southeast Asia. Am. J. Hum. Genet. 80, 29–43(2007).

26 Soares, P., Trejaut, J. A., Loo, J.-H., Hill, C., Mormina, M., Lee, C.-L. et al. Climatechange and postglacial human dispersals in Southeast Asia. Mol. Biol. Evol. 25,

1209–1218 (2008).27 Cox, M. P., Karafet, T. M., Lansing, J. S., Sudoyo, H. & Hammer, M. F. Autosomal and

X-linked single nucleotide polymorphisms reveal a steep Asian-Melanesian ancestrycline in eastern Indonesia and a sex bias in admixture rates. Proc. R. Soc. B 277,

1589–1596 (2010).28 HUGO Pan-Asian SNP Consortium. Mapping human genetic diversity in Asia. Science

326, 1541–1545 (2009).29 Wollstein, A., Lao, O., Becker, C., Brauer, S., Trent, R. J., Nurnberg, P. et al.

Demographic history of Oceania inferred from genome-wide data. Curr. Biol. 20,

1983–1992 (2010).30 Yao, Y. G., Kong, Q. P., Bandelt, H. J., Kivisild, T. & Zhang, Y. P. Phylogeographic

differentiation of mitochondrial DNA in Han Chinese. Am. J. Hum. Genet. 70,

635–651 (2002).31 Allard, M. W., Wilson, M. R., Monson, K. L. & Budowle, B. Control region sequences for

East Asian individuals in the Scientific Working Group on DNA Analysis Methodsforensic mtDNA data set. Leg. Med. 6, 11–24 (2004).

32 Li, H., Cai, X., Winograd-Cort, E. R., Wen, B., Cheng, X., Qin, Z. et al. MitochondrialDNA diversity and population differentiation in southern East Asia. Am. J. Phys.Anthropol. 134, 481–488 (2007).

33 Hill, C., Soares, P., Mormina, M., Macaulay, V., Meehan, W., Blackburn, J. et al.Phylogeography and ethnogenesis of aboriginal southeast Asians. Mol. Biol. Evol. 23,

2480–2491 (2006).34 Trejaut, J. A., Kivisild, T., Loo, J. H., Lee, C. L., He, C. L., Hsu, C. J. et al. Traces of

archaic mitochondrial lineages persist in Austronesian-speaking Formosan populations.PLoS Biol. 3, e247 (2005).

35 Friedlaender, J. S., Friedlaender, F. R., Hodgson, J. A., Stoltz, M., Koki, G., Horvat, G.et al. Melanesian mtDNA complexity. PLoS ONE 2, e248 (2007).

36 Tabbada, K. A., Trejaut, J., Loo, J.-H., Chen, Y.-M., Lin, M., Mirazon-Lahr, M. et al.Philippine mitochondrial DNA diversity: a populated viaduct between Taiwan andIndonesia? Mol. Biol. Evol. 27, 21–31 (2010).

37 Miller, S. A., Dykes, D. D. & Polesky, H. F. A simple salting out procedure for extractingDNA from human nucleated cells. Nucleic Acids Res. 16, 1215 (1988).

38 Andrews, R. M., Kubacka, I., Chinnery, P. F., Lightowlers, R. N., Turnbull, D. M. &Howell, N. Reanalysis and revision of the Cambridge reference sequence for humanmitochondrial DNA. Nat. Genet. 23, 147 (1999).

39 Tanaka, M., Cabrera, V. M., Gonzalez, A. M., Larruga, J. M., Takeyasu, T., Fuku, N.et al. Mitochondrial genome variation in eastern Asia and the peopling of Japan.Genome Res. 14, 1832–1850 (2004).

40 Kong, Q. P., Yao, Y. G., Sun, C., Bandelt, H. J., Zhu, C. L. & Zhang, Y. P. Phylogeny ofeast Asian mitochondrial DNA lineages inferred from complete sequences. Am. J.Hum. Genet. 73, 671–676 (2003).

41 Friedlaender, J., Gentz, F., Friedlaender, F., Kaestle, F., Schurr, T., Koki, G. et al.Mitochondrial genetic diversity and its determinants in Island Melanesia. In PapuanPasts: Cultural, Linguistic and Biological Histories of Papuan-Speaking Peoples (edsPawley, A., Attenborough, R., Golson, J. & Hide, R.) 693–716 (Pacific Linguistics,Canberra, 2005).

42 van Oven, M. Revision of the mtDNA tree and corresponding haplogroup nomenclature.PNAS 107, E38–E39 (2010).

43 Excoffier, L. & Lischer, H. E. L. Arlequin suite ver 3.5: a new series of programs toperform population genetics analyses under Linux and Windows. Mol. Ecol. Res. 10,

564–567 (2010).44 Slatkin, M. Inbreeding coefficients and coalescence times. Genet. Res. Camb. 58,

167–175 (1991).45 Slatkin, M. A measure of population subdivision based on microsatellite allele

frequencies. Genetics 139, 457–462 (1995).46 R Development Core Team. R: A Language and Environment for Statistical Computing

(2012). http://www.r-project.org/.47 Bandelt, H. J., Forster, P. & Rohl, A. Median-joining networks for inferring intraspecific

phylogenies. Mol. Biol. Evol. 16, 37–48 (1999).48 Forster, P., Harding, R., Torroni, A. & Bandelt, H.-J. Origin and evolution of

native American mtDNA variation: a reappraisal. Am. J. Hum. Genet. 59, 935–945(1996).

49 Soares, P., Ermini, L., Thomson, N., Mormina, M., Rito, T., Rohl, A. et al. Correcting forpurifying selection: an improved human mitochondrial molecular clock. Am. J. Hum.Genet. 84, 740–759 (2009).

50 Cox, M. P. Accuracy of molecular dating with the rho statistic: deviations fromcoalescent expectations under a range of demographic models. Hum. Biol. 80,

335–357 (2008).51 Hedrick, P. W. A standardized genetic differentiation measure. Evolution 59,

1633–1638 (2005).52 van Oven, M., Hammerle, J. M., van Schoor, M., Kushnick, G., Pennekamp, P., Zega, I.

et al. Unexpected island effects at an extreme: reduced Y-chromosome and mitochon-drial DNA diversity in Nias. Mol. Biol. Evol. 28, 1349–1361 (2011).

53 Cox, M. P., Nelson, M. G., Tumonggor, M. K., Ricaut, F. X. & Sudoyo, H. A smallcohort of Island Southeast Asian women founded Madagascar. Proc. R. Soc. B 279,

2761–2768 (2012).54 Razafindrazaka, H., Ricaut, F.-X., Cox, M. P., Mormina, M., Dugoujon, J.-M.,

Randriamarolaza, L. P. et al. Complete mitochondrial DNA sequences provide newinsights into the Polynesian motif and the peopling of Madagascar. Eur. J. Hum. Genet.18, 575–581 (2010).

55 Macaulay, V., Hill, C., Achilli, A., Rengo, C., Clarke, D., Meehan, W. et al. Single, rapidcoastal settlement of Asia revealed by analysis of complete mitochondrial genomes.Science 308, 1034–1036 (2005).

56 Dupuy, B. M., Stenersen, M., Egeland, T. & Olaisen, B. Y-chromosomal microsatellitemutation rates: differences in mutation rate between and within loci. Hum. Mutat. 23,

117–124 (2004).57 Fenner, J. N. Cross-cultural estimation of the human generation interval for use in

genetics-based population divergence studies. Am. J. Phys. Anthropol. 128, 415–423(2005).

58 Gusmao, L., Sanchez-Diz, P., Calafell, F., Martın, P., Alonso, C. A., Alvarez-Fernandez,F. et al. Mutation rates at Y chromosome specific microsatellites. Hum. Mutat. 26,

520–528 (2005).59 Zhivotovsky, L. A., Underhill, P. A., Cinnioglu, C., Kayser, M., Morar, B., Kivisild, T.

et al. The effective mutation rate at Y chromosome short tandem repeats, withapplication to human population-divergence time. Am. J. Hum. Genet. 74, 50–61(2004).

60 Lansing, J. S., Karafet, T. M., Schoenfelder, J. & Hammer, M. F. A DNA signature forthe expansion of irrigation in Bali? In Past Human Migrations in East Asia: MatchingArchaeology, Linguistics and Genetics (eds Sanchez-Mazas, A., Blench, R., Ross, M.D., Peiros, I. & Lin, M.) 377–395 (Routledge, London, 2008).

61 Blust, R. The Austronesian Languages. Research School of Pacific and Asian Studies.(Australian National University, Canberra, Australia, 2009).

62 Oppenheimer, S. & Richards, M. Fast trains, slow boats, and the ancestry of thePolynesian Islanders. Sci. Prog. 84, 157–181 (2001).

63 Jinam, T. A., Hong, L.-C., Phipps, M. E., Stoneking, M., Ameen, M., Edo, J. et al.Evolutionary history of continental southeast Asians: ‘Early train’ hypothesis basedon genetic analysis of mitochondrial and autosomal DNA data. Mol. Biol. Evol 29,

3513–3527 (2012).64 Soares, P., Rito, T., Trejaut, J., Mormina, M., Hill, C., Tinkler-Hundal, E. et al. Ancient

voyaging and Polynesian origins. Am. J. Hum. Genet. 88, 239–247 (2011).65 Forster, P., Torroni, A., Renfrew, C. & Rohl, A. Phylogenetic star contraction applied to

Asian and Papuan mtDNA evolution. Mol. Biol. Evol. 18, 1864–1881 (2001).66 Ross, M. Pronouns as a preliminary diagnostic of grouping Papuan languages. In

Papuan Pasts: Cultural, Linguistic and Biological Histories of Papuan-SpeakingPeoples (eds Pawley, A., Attenborough, R., Golson, J. & Hide, R.) 15–65 (PacificLinguistics, Canberra, 2005).

67 Xu, S., Pugach, I., Stoneking, M., Kayser, M., Jin, L. & the HUGO Pan-Asian SNPConsortium. Genetic dating indicates that the Asian-Papuan admixture through EasternIndonesia corresponds to the Austronesian expansion. Proc. Natl Acad. Sci. USA 109,

4574–4579 (2012).68 Gray, R. D. & Jordan, F. M. Language trees support the express-train sequence of

Austronesian expansion. Nature 405, 1052–1054 (2000).69 Hage, P. & Marck, J. Matrilineality and the Melanesian origin of Polynesian Y

chromosomes. Curr. Anthropol. 44, S121–S127 (2003).70 Lansing, S. J., Cox, M. P., de Vet, T. A., Downey, S. S., Hallmark, B. & Sudoyo, H. An

ongoing Austronesian expansion in Island Southeast Asia. J. Anthropol. Archaeol. 30,

262–272 (2011).71 Tibbetts, G. A Study of the Arabic Texts Containing Material on South-East Asia (E. J.

Brill, Leiden, 1979).

Supplementary Information accompanies the paper on Journal of Human Genetics website (http://www.nature.com/jhg)


173


http://www.r-project.org/

http://www.nature.com/jhg

The Indonesian archipelago: an ancient genetic highway linking Asia and the Pacific

Documents